Tran Le
Member Of Technical Staff at Fireworks AI
San Francisco, California, United States
Summary
🤩
Rockstar🎓
Top SchoolTran Le is a Member of Technical Staff and machine learning engineer with 7 years of experience building scalable training and inference systems for recommender and LLM workloads in the Bay Area. A Grinnell-educated mathematician and computer scientist originally from Vietnam, he has shipped production features at Meta across TorchRec, privacy-preserving federated learning, and feed ranking, and now works on post-training platforms at Fireworks AI. He contributes to PyTorch's TorchRec—helping improve distributed planning and storage estimation for large-scale recommendation systems—demonstrating deep systems-level expertise in sparsity and parallelism. Comfortable across research-adjacent and production environments, Tran blends strong backend engineering with applied ML at scale. Outside core engineering, he founded a literacy-support project in Vietnam, showing an early commitment to social impact and pragmatic leadership.
7 years of coding experience
7 years of employment as a software developer
High School Diploma, English Language and Literature/Letters, High School Diploma, English Language and Literature/Letters at Quoc Hoc High School For the Gifted StudentsBachelor's degree, Mathematics and Computer Science, Bachelor's degree, Mathematics and Computer Science at Grinnell College
English, Vietnamese
Github Skills (9)
pytorch10
distributed-systems10
recommendation-system10
python10
deep-learning9
cuda8
sharding8
machine-learning7
compiler6
Programming languages (4)
Github contributions (5)
pytorch/torchrec
Aug 2022 - Jan 2023
Pytorch domain library for recommendation systems
Role in this project:
Back-end Developer
Back-end DeveloperContributions:2 reviews, 15 commits, 35 PRs in 4 months
Contributions summary:Tran primarily contributed to the TorchRec library, focusing on the development and refinement of the distributed planner and storage estimation functionalities. Their work included adding features to planner stats logging, enabling the consideration of multiple optimizer classes in storage estimation, and addressing issues related to storage overestimation. Furthermore, they made changes to the code and test files.
cudapytorchrecommendation-systemsdeep-learninggpu
lequytra/torchrec
Aug 2022 - May 2024
Pytorch domain library for recommendation systems
Contributions:167 pushes, 24 branches in 1 year 8 months
pytorchrecommendation-systemsdataminingrecommendationdomain
Find and Hire Top DevelopersWe’ve analyzed the programming source code of over 60 million software developers on GitHub and scored them by 50,000 skills. Sign-up on Prog,AI to search for software developers.